Clinical Psychological Science 2013 Neshat Doost 84 92

8/13/2019 Clinical Psychological Science 2013 Neshat Doost 84 92

1/10

http://cpx.sagepub.com/Clinical Psychological Science

http://cpx.sagepub.com/content/1/1/84The online version of this article can be found at:

DOI: 10.1177/2167702612454613

2013 1: 84 originally published online 7 September 2012Clinical Psychological ScienceJobson

mid Taher Neshat-Doost, Tim Dalgleish, William Yule, Mehrdad Kalantari, Sayed Jafar Ahmadi, Atle Dyregrov andDepression Translated From Basic Science

Enhancing Autobiographical Memory Specificity Through Cognitive Training: An Intervention fo

Published by:

http://www.sagepublications.com

On behalf of:

Association for Psychological Science

can be found at:Clinical Psychological ScienceAdditional services and information for

http://cpx.sagepub.com/cgi/alertsEmail Alerts:

http://cpx.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

What is This?

-Sep 7, 2012OnlineFirst Version of Record

- Dec 14, 2012Version of Record>>

at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from at Alexandru Ioan Cuza on October 31, 2013cpx.sagepub.comDownloaded from
http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/content/1/1/84http://cpx.sagepub.com/content/1/1/84http://www.sagepublications.com/http://www.sagepublications.com/http://www.psychologicalscience.org/http://cpx.sagepub.com/cgi/alertshttp://cpx.sagepub.com/cgi/alertshttp://cpx.sagepub.com/subscriptionshttp://cpx.sagepub.com/subscriptionshttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsPermissions.navhttp://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://cpx.sagepub.com/content/early/2012/09/07/2167702612454613.full.pdfhttp://cpx.sagepub.com/content/early/2012/09/07/2167702612454613.full.pdfhttp://cpx.sagepub.com/content/1/1/84.full.pdfhttp://cpx.sagepub.com/content/1/1/84.full.pdfhttp://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://cpx.sagepub.com/http://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://cpx.sagepub.com/content/early/2012/09/07/2167702612454613.full.pdfhttp://cpx.sagepub.com/content/1/1/84.full.pdfhttp://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsReprints.navhttp://cpx.sagepub.com/subscriptionshttp://cpx.sagepub.com/cgi/alertshttp://www.psychologicalscience.org/http://www.sagepublications.com/http://cpx.sagepub.com/content/1/1/84http://cpx.sagepub.com/


2/10

Clinical Psychological Science

1(1) 8492 The Author(s) 2013

Reprints and permission:

sagepub.com/journalsPermissions.navDOI: 10.1177/2167702612454613

http://cpx.sagepub.com

Reduced autobiographical memory specificity (AMS) is con-

sidered a cognitive marker and predictor of the course of depres-

sion (Sumner, Griffith, & Mineka, 2010). That is, individuals

with depression, those in remission, and formerly depressed

individuals all have significant difficulties providing specific

memories (Williams et al., 2007). There is now substantive evi-

dence that reduced AMS has important implications for the

functionality of everyday cognition. Reduced AMS has beenfound to be associated with impaired everyday and interper-

sonal problem solving (Raes et al., 2005; Sutherland &

Bryant, 2008), difficulties in imagining specific events in the

future (Williams et al., 1996), rumination (Williams, 1996), and

a lack of exposure to negative memories and associated nega-

tive emotionswhich, although unpleasant in the short term, is

associated with well-being in the longer term (Raes, Williams,

& Hermans, 2009). Consequently, these difficulties potentially

result in ongoing negative social encounters, diminished posi-

tive social reinforcement and self-efficacy, feelings of hopeless-

ness, and depressed mood (Raes et al., 2009). In light of these

relationships, it is not surprising that reduced AMS has been

found to predict subsequent major depression at 6 months, over

and above what could be predicted from initial symptom sever-

ity (Kleim & Ehlers, 2008). Importantly, research suggests that

reduced AMS is not a fixed feature of an individuals mnemonic

style (Yeung, Dalgleish, Golden, & Schartau, 2006); it can be

modified (Williams, Teasdale, Segal, & Soulsby, 2000). There-

fore, it is proposed that clinical efforts to improve AMS shouldhave positive clinical outcomes for patients who are depressed

(Raes et al., 2009).

Preliminary research has investigated whether a brief train-

ing program targeting AMS can mitigate this known vulnera-

bility factor for depression and whether these improvements

Corresponding Author:

Laura Jobson, Department of Psychological Sciences, Norwich Medical

School, University of East Anglia, Norwich Research Park, Norwich,

Norfolk, United Kingdom NR4 7TJ

E-mail: [email protected]

Enhancing Autobiographical MemorySpecificity Through Cognitive Training:

An Intervention for Depression TranslatedFrom Basic Science

Hamid Taher Neshat-Doost1, Tim Dalgleish2, William Yule3,

Mehrdad Kalantari1, Sayed Jafar Ahmadi1, Atle Dyregrov3,

and Laura Jobson2,41University of Isfahan, Islamic Republic of Iran; 2Medical Research Council Cognition and Brain Sciences

Unit, Cambridge, England; 3Children and War Foundation, Bergen, Norway; and 4University

of East Anglia, Norwich, England

Abstract

The objective of this study was to investigate the efficacy of memory specificity training (MEST) on autobiographical memory

recall and depression. Afghan adolescents with depression were randomly assigned to a MEST group or to a control group.At baseline, both groups completed Persian versions of the Autobiographical Memory Test (AMT) and the Mood and

Feelings Questionnaire (MFQ). The MEST group then had five weekly group sessions of MEST. The control group had no

additional contact. The AMT and MFQ were then readministered to all participants, and the MFQ was readministered at

2-month follow-up. The MEST group retrieved a higher proportion of specific memories following training and had lowerlevels of depression at 2-month follow-up than did the control group. Change in memory specificity predicted follow-up

depression over and above baseline depression and mediated the relationship between receipt of MEST and reduction in

later depression. The results suggest that MEST can improve autobiographical memory performance and drive subsequent

reduction in depression symptoms.

Keywords

memory specificity training, autobiographical memory, depression, adolescence

Received 4/5/12; Revision accepted 6/8/12

Brief Empirical Report


3/10

Intervention for Depression 85

are accompanied by improvements in the variables assumed to

mediate the influence of AMS on the course of depression

(Raes et al., 2009). Raes et al. evaluated the effect of a group-

based standalone training program (memory specificity train-

ing; MEST) comprising four weekly 1-hour sessions with 10

depressed female inpatients. They found that participants

retrieval styles became significantly more specific and thatobserved improvements in AMS were significantly associated

with improvements in rumination, cognitive avoidance, and

problem-solving skills. As the researchers acknowledge, this

was a preliminary study hampered by the absence of a control

group, by a relatively high dropout rate, and because there was

no follow-up period. This latter issue is particularly important

because, theoretically, one would expect that any advantages in

terms of depression symptoms accrued by the reduced specific-

ity of memories would emerge some weeks following the mem-

ory interventiona point that the researchers emphasize and

that led to their excluding any analysis of depressive symptoms

in their report (Raes et al., 2009; Williams et al., 2007). These

limitations notwithstanding, this initial study supports the theo-retical hypothesis that it is possible to modify AMS and vali-

dates a training program to achieve this goal. The data suggest

that MEST may be a promising program that improves a core

cognitive component of depression (Raes et al., 2009).

The present study takes this research to the next stage. Three

key questions are addressed. First, in a blinded, randomized

design, can participants in the MEST group increase AMS

compared with those in the control group? Second, does MEST

lead to any improvement in depressive symptomatology either

immediately following training or, more likely, at follow-up?

Third, do increases in AMS over the course of training mediate

any improvements in depression at follow-up? This latter ques-tion is particularly important, as it speaks to the theoretically

predicted causal relationship between AMS enhancement and

depression reduction (Williams et al., 2007).

To explore these questions, we recruited bereaved adoles-

cents who were currently experiencing significant symptoms

of depression and who were recent refugee immigrants from a

war zone. We know that adolescents with depression also pres-

ent with reduced AMS (Kuyken & Dalgleish, 2011), which is

of clinical importance given that adolescence is a central

developmental period in terms of the etiology and long-term

prognosis of depression (Kuyken & Dalgleish, 2011). We also

know that reduced AMS is associated with the experience of

trauma (Dalgleish, Rolfe, Golden, Dunn, & Barnard, 2008)and with difficulties following bereavement (Golden,

Dalgleish, & Mackintosh, 2007). This sample is therefore an

ideal group to address the current research questions.

Method

Participants

We used data from Raes et al. (2009) as the basis for a power

calculation. In their study, MEST produced a change in the

proportional measure of AMS of 1.1 standard deviation units.

Based on this effect size, and assuming no change in AMS in

our control group, we calculated that a sample size of 22 (11 per

group) would provide 80% power, with a directional alpha of

.05, to detect a similar improvement in AMS in the present

study. We approached a charity school for adolescents in Qhom,

Iran, to collaborate in providing participants for the study. Sev-enty adolescents consented to be assessed, all of whom had lost

their fathers due to the war in Afghanistan and had subsequently

immigrated to Iran as refugees. Following Raes et al.s proce-

dure, participants were selected if they were experiencing sig-

nificant levels of depressive symptomatology. Specifically, to

be included in the training study, participants had to score

higher than 27 on the Persian version of the Mood and Feeling

Questionnaire (pMFQ). The Mood and Feeling Questionnaire

(MFQ) is a well-established questionnaire intended for use with

adolescents; it covers a broad range of depression symptoms.

The cutoff of 27 was selected because research indicates that

adolescents who score above 27 on the MFQ meet the screen-

ing criteria for major depression disorder; these results holdfor the pMFQ (Neshat-Doost, Nouri, Molavi, Kalantari, &

Mehrabi, 2006; Park, Goodyer, & Teasdale, 2002; Wood, Kroll,

Moore, & Harrington, 1995). We decided not to exclusively

select participants who met the diagnosis for depression,

because reduced AMS has been found to be a transdiagnostic

impairment and to be present in those who are in remission or

have recovered from depression; also, the MFQ has been found

to be sensitive to detecting high-level depression symptoms.

Twenty-three adolescents met this inclusion criterion on the

pMFQ and were randomized1to either MEST (n= 12) or the

control condition (n= 11). The mean number of years partici-

pants had been in Iran was 9.51 (SD= 3.28). There were 11girls and 12 boys. The mean age of the participants was 14.88

years (SD= 1.89). The mean number of years of education was

6.84 (SD= 2.08).

Materials

Autobiographical Memory Task (AMT). The AMT

(Williams & Broadbent, 1986 ) is the gold-standard test of

memory specificity, and the version presented here was identi-

cal to that prototypically employed in the existing research.

The AMT has adequate psychometric properties (Griffith et

al., in press). In the present study, participants were presented

with 18 Persian cue words consisting of positive (e.g., party,

), negative (e.g., accident, ), and neutral (e.g.,

class, ) words. These words were selected based on apilot study in which the positive and negative emotionality of

several hundred Persian words were rated by 60 raters. Three

psychologists then selected six positive, six negative, and six

neutral words with the highest emotionality ratings that were

also balanced for frequency of use and length (Griffith et al.,

2009). Before use in the current study, these words were tested

for feasibility as memory cues with 30 Afghan adolescents liv-

ing in Isfahan.


4/10

86 Neshat-Doost et al.

The cue words were presented in written form, each on an

A4 card, in a separate random order for each participant. Par-

ticipants were instructed to read each word aloud, and they

were then given 30 s in each case to retrieve a specific mem-

ory. Participants were told that the memory they recalled could

be an important or trivial event that happened recently or a

long time ago, but that it should be something that happened ata particular time on a particular day. Examples of acceptable

and unacceptable responses were given. Three practice cues

were given (gigantic, bag, and homework). Generated memo-

ries were tape recorded for coding. Specific memories were

defined as events that lasted for a day or less. Nonspecific

memories were coded as extended (events that lasted for lon-

ger periods of time) or categoric (events that occurred repeat-

edly over a period of time) memories, in line with the previous

literature (Williams et al., 2007); these data are presented in

the Results section. If a participant failed to recall a memory or

talked about things that were not memories, the response was

classified as no memory. The raters were blind to the studys

aims and to the group allocation of participants. A secondassessor rated 20% of the memories. Interrater reliability for

specificity and nonspecificity was good (= .92). Research

has found that, in the absence of any manipulation or interven-

tion, there is consistency and stability in scores on the AMT

across time; this is consistent with the notion of AMS as a

stable trait variable (Griffith et al., in press). Therefore, as in

previous research examining change in AMS (Sutherland &

Bryant, 2008), parallel forms of the AMT were not employed.

The inclusion of a control group also ensured that in the

unlikely event of any practice effects in the experimental

group, such changes would be taken account of in the

analyses.

Adult Reading Test (ART).The Persian version (Haghshenas,

Farashbandi, Mani, & Tahmasbi, 2001) of the ART (Brooks,

Everatt, & Fidler, 2004) assesses the literacy skills of adoles-

cents and adults in Persian. This was included to assess the

comparability of the groups in terms of verbal ability, which

can influence performance on the AMT (Williams et al., 2007).

It has been standardized for Persian speakers, and the psycho-

metric properties have been found to be good (Haghshenas

et al., 2001).

Mood and Feeling Questionnaire (MFQ).We used the Per-

sian version (Neshat-Doost et al., 2006) of the MFQ (Angold,Costello, Pickles, & Winder, 1987), which is a 32-item ques-

tionnaire based on criteria for depression. It consists of a series

of descriptive phrases regarding how the participant has

been feeling or acting, and the adolescent rates whether the

phrase was descriptive of most of the time, sometimes, or not

at allin the past 2 weeks. The reliability and validity of the

Persian measure has been found to be good (Neshat-Doost

et al., 2006). In the current study, internal consistency was

high ( = .92).

Revised Childrens Manifest Anxiety Scale (RCMAS).To

assess anxiety and ensure comparability across groups, the

Persian version of the RCMAS (Reynolds & Richmond, 1978)

was used because it measures the level and nature of anxiety

as experienced by adolescents and has good psychometric

properties (Taghavi & Alishahi, 2004). In the current study,

internal consistency was good ( = .79).

Impact of Event Scale (IES).The IES (Horowitz, Wilner, &

Alvarez, 1979) is a 15-item self-report questionnaire measur-

ing the severity of symptoms of intrusion and avoidance

around a specified eventin this case, paternal bereavement.

The Persian version of the IES was used (Moradi, Herlihy,

Yasseri, Turner, & Dalgleish, 2008). In the current study, inter-

nal consistency was good ( = .83). We included the IES to

provide a measure of postevent symptomatology at baseline so

that we could ensure comparability across the two groups.

MEST

MEST was conducted by trained senior clinical psychologists,Hamid Taher Neshat-Doost and Sayed Jafar Ahmadi. The

training package consisted of five weekly 80-min group ses-

sions in one of the classrooms. All words used in the training

were different from the words used in the AMT.

Session 1 included the aims and outline of MEST. At the

beginning of this session, the definition and importance of key

terms, such as autobiographical memory and recall, and the

three types of autobiographical memory (specific, extended,

and categoric) were explained. Examples of each of the recall

types were written on a whiteboard. At the end of the session,

participants were given 10 cue words and were instructed to

write down specific, extended, and categoric memories foreach of the words during the next week as homework. Each

participant was provided with a pen and notebook to use for

summarizing sessions and for homework tasks.

Session 2 commenced with a review of Session 1 and

homework. The group then was prompted to give positive cue

words. These words were written on the whiteboard, and par-

ticipants were asked to provide a specific memory for each of

the cue words. If one of the participants recalled a nonspecific

memory, the type of the memory (i.e., categoric or extended)

was clarified, and the participants were prompted to recall spe-

cific memories. At the end of the session, 10 positive cue

words were given and participants were asked to provide one

specific memory for each of the words as homework.Sessions 3 and 4 were very similar to Session 2, as were the

associated homework tasks. The only difference was the con-

tent of the cue words. The cue words were negative for Ses-

sion 3 and neutral for Session 4.

Session 5 was again similar to previous sessions. The dis-

tinction between different types of autobiographical memory

recall was practiced with positive, negative, and neutral cue

words, and the difference between a memory and a nonmem-

ory was clarified.


5/10

Intervention for Depression 87

Procedure

Ethical approval was obtained from the University of Isfahan.

An information sheet and a consent form were given to partici-

pants and their parents or guardians. Participants were tested

individually in a classroom by researchers blind to group status

on three occasions: at pretraining, posttraining, and 2-monthposttraining follow-up. The pretraining assessment consisted of

the AMT, ART, MFQ, IES, and RCMAS. Following this, we

randomly allocated participants to either MEST or the control

group (which had no additional contact). The AMT, RCMAS,

and MFQ were readministered at posttraining, and the RCMAS

and MFQ were again readministered 2 months later. Figure 1

summarizes the trial CONSORT (consolidated standards of

reporting trials) diagram.

Results

Participant characteristics

Participant characteristics and group comparisons are pre-

sented in Table 1. The groups did not differ significantly on

demographic or baseline measures.

Analysis Including Pre- and Postintervention

Data (n = 12)

Analysis Including Follow-Up Data (n = 11)

Lost to Follow-Up (n = 0)

Allocated to Control (n = 11)

ExcludedScored Less Than 27 on theMFQ (n = 47)

Assessed for Eligibility (n = 70)

Analysis Including Pre- and Postintervention

Data (n = 11).

Analysis Including Follow-Up Data (n = 8)

Allocation

Follow-Up

Post-Training

Completed Follow-Up (n = 11)

Lost to Follow-Up (moved to another city inIran) (n = 1)

Completed Follow-Up (n = 8)

Lost to Follow-Up (moved to anothercountry n = 2; or city in Iran n = 1)

Analysis

Allocated to MEST Intervention (n = 12)

Received Allocated Intervention (n = 12)

Completed MEST Intervention (n = 12)

Lost to Follow-Up (n = 0)

Randomized (n = 23)

Enrollment

Fig. 1. CONSORT (consolidated standards of reporting trials) diagram of the progress through the phases of therandomized trial.

Note: MFQ = Mood and Feeling Questionnaire; MEST = memory specificity training.


6/10

88 Neshat-Doost et al.

Research Question 1: Specificity

Table 1 shows the mean proportion of specific memories

retrieved across the two groups at pre- and posttraining.2A 2

(group: MEST, control) 2 (time: pretraining, posttraining)

mixed analysis of variance (ANOVA) was used with the pro-

portion of specific memories as the dependent variable. There

was a significant Group Time interaction, F(1, 21) = 9.84,

p< .01, p2= .32. Table 1 shows that although there was no

significant difference between the groups at pretraining (d

Documents

Clinical Psychological Science 2013 Neshat Doost 84 92